This paper presents the systematic synthesis of a fairly complex digitalcircuit and its CPLD implementation as an assemblage of communicatingasynchronous sequential circuits. The example, a VMEbus controller, waschosen because it has to control concurrent processes and to arbitrateconflicting requests.
It is well known that the performance of mobile radio systems can be significantly enhanced by the application of adaptive antennas which consist of multi-element antenna arrays plus signal processing circuitry. In the thesis the utilization of such antennas as receive antennas in the uplink of mobile radio air interfaces of the type TD-CDMA is studied. Especially, the incorporation of covariance matrices of the received interference signals into the signal processing algorithms is investigated with a view to improve the system performance as compared to state of the art adaptive antenna technology. These covariance matrices implicitly contain information on the directions of incidence of the interference signals, and this information may be exploited to reduce the effective interference power when processing the signals received by the array elements. As a basis for the investigations, first directional models of the mobile radio channels and of the interference impinging at the receiver are developed, which can be implemented on the computer at low cost. These channel models cover both outdoor and indoor environments. They are partly based on measured channel impulse responses and, therefore, allow a description of the mobile radio channels which comes sufficiently close to reality. Concerning the interference models, two cases are considered. In the one case, the interference signals arriving from different directions are correlated, and in the other case these signals are uncorrelated. After a visualization of the potential of adaptive receive antennas, data detection and channel estimation schemes for the TD-CDMA uplink are presented, which rely on such antennas under the consideration of interference covariance matrices. Of special interest is the detection scheme MSJD (Multi Step Joint Detection), which is a novel iterative approach to multi-user detection. Concerning channel estimation, the incorporation of the knowledge of the interference covariance matrix and of the correlation matrix of the channel impulse responses is enabled by an MMSE (Minimum Mean Square Error) based channel estimator. The presented signal processing concepts using covariance matrices for channel estimation and data detection are merged in order to form entire receiver structures. Important tasks to be fulfilled in such receivers are the estimation of the interference covariance matrices and the reconstruction of the received desired signals. These reconstructions are required when applying MSJD in data detection. The considered receiver structures are implemented on the computer in order to enable system simulations. The obtained simulation results show that the developed schemes are very promising in cases, where the impinging interference is highly directional, whereas in cases with the interference directions being more homogeneously distributed over the azimuth the consideration of the interference covariance matrices is of only limited benefit. The thesis can serve as a basis for practical system implementations.
In conventional radio communication systems, the system design generally starts from the transmitter (Tx), i.e. the signal processing algorithm in the transmitter is a priori selected, and then the signal processing algorithm in the receiver is a posteriori determined to obtain the corresponding data estimate. Therefore, in these conventional communication systems, the transmitter can be considered the master and the receiver can be considered the slave. Consequently, such systems can be termed transmitter (Tx) oriented. In the case of Tx orientation, the a priori selected transmitter algorithm can be chosen with a view to arrive at particularly simple transmitter implementations. This advantage has to be countervailed by a higher implementation complexity of the a posteriori determined receiver algorithm. Opposed to the conventional scheme of Tx orientation, the design of communication systems can alternatively start from the receiver (Rx). Then, the signal processing algorithm in the receiver is a priori determined, and the transmitter algorithm results a posteriori. Such an unconventional approach to system design can be termed receiver (Rx) oriented. In the case of Rx orientation, the receiver algorithm can be a priori selected in such a way that the receiver complexity is minimum, and the a posteriori determined transmitter has to tolerate more implementation complexity. In practical communication systems the implementation complexity corresponds to the weight, volume, cost etc of the equipment. Therefore, the complexity is an important aspect which should be taken into account, when building practical communication systems. In mobile radio communication systems, the complexity of the mobile terminals (MTs) should be as low as possible, whereas more complicated implementations can be tolerated in the base station (BS). Having in mind the above mentioned complexity features of the rationales Tx orientation and Rx orientation, this means that in the uplink (UL), i.e. in the radio link from the MT to the BS, the quasi natural choice would be Tx orientation, which leads to low cost transmitters at the MTs, whereas in the downlink (DL), i.e. in the radio link from the BS to the MTs, the rationale Rx orientation would be the favorite alternative, because this results in simple receivers at the MTs. Mobile radio downlinks with the rationale Rx orientation are considered in the thesis. Modern mobile radio communication systems are cellular systems, in which both the intracell and intercell interferences exist. These interferences are the limiting factors for the performance of mobile radio systems. The intracell interference can be eliminated or at least reduced by joint signal processing with consideration of all the signals in the considered cell. However such joint signal processing is not feasible for the elimination of intercell interference in practical systems. Knowing that the detrimental effect of intercell interference grows with its average energy, the transmit energy radiated from the transmitter should be as low as possible to keep the intercell interference low. Low transmit energy is required also with respect to the growing electro-phobia of the public. The transmit energy reduction for multi-user mobile radio downlinks by the rationale Rx orientation is dealt with in the thesis. Among the questions still open in this research area, two questions of major importance are considered here. MIMO is an important feature with respect to the transmit power reduction of mobile radio systems. Therefore, first questionconcerns the linear Rx oriented transmission schemes combined with MIMO antenna structures. The investigations of the MIMO benefit on the linear Rx oriented transmission schemes are studied in the thesis. Utilization of unconventional multiply connected quantization schemes at the receiver has also great potential to reduce the transmit energy. Therefore, the second question considers the designing of non-linear Rx oriented transmission schemes combined with multiply connected quantization schemes.
Rapid growth in sensors and sensor technology introduces variety of products to the market. The increasing number of available sensor concepts and implementations demands more versatile sensor electronics and signal conditioning. Nowadays signal conditioning for the available spectrum of sensors is becoming more and more challenging. Moreover, developing a sensor signal conditioning ASIC is a function of cost, area, and robustness to maintain signal integrity. Field programmable analog approaches and the recent evolvable hardware approaches offer partial solution for advanced compensation as well as for rapid prototyping. The recent research field of evolutionary concepts focuses predominantly on digital and is at its advancement stage in analog domain. Thus, the main research goal is to combine the ever increasing industrial demand for sensor signal conditioning with evolutionary concepts and dynamically reconfigurable matched analog arrays implemented in main stream Complementary Metal Oxide Semiconductors (CMOS) technologies to yield an intelligent and smart sensor system with acceptable fault tolerance and the so called self-x features, such as self-monitoring, self-repairing and self-trimming. For this aim, the work suggests and progresses towards a novel, time continuous and dynamically reconfigurable signal conditioning hardware platform suitable to support variety of sensors. The state-of-the-art has been investigated with regard to existing programmable/reconfigurable analog devices and the common industrial application scenario and circuits, in particular including resource and sizing analysis for proper motivation of design decisions. The pursued intermediate granular level approach called as Field Programmable Medium-granular mixed signal Array (FPMA) offers flexibility, trimming and rapid prototyping capabilities. The proposed approach targets at the investigation of industrial applicability of evolvable hardware concepts and to merge it with reconfigurable or programmable analog concepts, and industrial electronics standards and needs for next generation robust and flexible sensor systems. The devised programmable sensor signal conditioning test chips, namely FPMA1/FPMA2, designed in 0.35 µm (C35B4) Austriamicrosystems, can be used as a single instance, off the shelf chip at the PCB level for conditioning or in the loop with dedicated software to inherit the aspired self-x features. The use of such self–x sensor system carries the promise of improved flexibility, better accuracy and reduced vulnerability to manufacturing deviations and drift. An embedded system, namely PHYTEC miniMODUL-515C was used to program and characterize the mixed-signal test chips in various feedback arrangements to answer some of the questions raised by the research goals. Wide range of established analog circuits, ranging from single output to fully differential amplifiers, was investigated at different hierarchical levels to realize circuits like instrumentation amplifier and filters. A more extensive design issues based on low-power like for e.g., sub-threshold design were investigated and a novel soft sleep mode idea was proposed. The bandwidth limitations observed in the state of the art fine granular approaches were enhanced by the proposed intermediate granular approach. The so designed sensor signal conditioning instrumentation amplifier was then compared to the commercially available products in the market like LT 1167, INA 125 and AD 8250. In an adaptive prototype, evolutionary approaches, in particular based on particle swarm optimization with multi-objectives, were just deployed to all the test samples of FPMA1/FMPA2 (15 each) to exhibit self-x properties and to recover from manufacturing variations and drift. The variations observed in the performance of the test samples were compensated through reconfiguration for the desired specification.
Channel estimation is of great importance in many wireless communication systems, since it influences the overall performance of a system significantly. Especially in multi-user and/or multi-antenna systems, i.e. generally in multi-branch systems, the requirements on channel estimation are very high, since the training signals or so called pilots that are used for channel estimation suffer from multiple access interference. Recently, in the context with such systems more and more attention is paid to concepts for joint channel estimation (JCE) which have the capability to eliminate the multiple access interference and also the interference between the channel coefficients. The performance of JCE can be evaluated in noise limited systems by the SNR degradation and in interference limited systems by the variation coefficient. Theoretical analysis carried out in this thesis verifies that both performance criteria are closely related to the patterns of the pilots used for JCE, no matter the signals are represented in the time domain or in the frequency domain. Optimum pilots like disjoint pilots, Walsh code based pilots or CAZAC code based pilots, whose constructions are described in this thesis, do not show any SNR degradation when being applied to multi-branch systems. It is shown that optimum pilots constructed in the time domain become optimum pilots in the frequency domain after a discrete Fourier transformation. Correspondingly, optimum pilots in the frequency domain become optimum pilots in the time domain after an inverse discrete Fourier transformation. However, even for optimum pilots different variation coefficients are obtained in interference limited systems. Furthermore, especially for OFDM-based transmission schemes the peak-to-average power ratio (PAPR) of the transmit signal is an important decision criteria for choosing the most suitable pilots. CAZAC code based pilots are the only pilots among the regarded pilot constructions that result in a PAPR of 0 dB for the transmit signal that origins in the transmitted pilots. When summarizing the analysis regarding the SNR degradation, the variation coefficient and the PAPR with respect to one single service area and considering the impact due to interference from other adjacent service areas that occur due to a certain choice of the pilots, one can conclude that CAZAC codes are the most suitable pilots for the application in JCE of multi-carrier multi-branch systems, especially in the case if CAZAC codes that origin in different mother codes are assigned to different adjacent service areas. The theoretical results of the thesis are verified by simulation results. The choice of the parameters for the frequency domain or time domain JCE is oriented towards the evaluated implementation complexity. According to the chosen parameterization of the regarded OFDM-based and FMT-based systems it is shown that a frequency domain JCE is the best choice for OFDM and a time domain JCE is the best choice for FMT applying CAZAC codes as pilots. The results of this thesis can be used as a basis for further theoretical research and also for future JCE implementation in wireless systems.
The work presented in this thesis discusses the thermal and power management of multi-core processors (MCPs) with both two dimensional (2D) package and there dimensional (3D) package chips. The power and thermal management/balancing is of increasing concern and is a technological challenge to the MCP development and will be a main performance bottleneck for the development of MCPs. This thesis develops optimal thermal and power management policies for MCPs. The system thermal behavior for both 2D package and 3D package chips is analyzed and mathematical models are developed. Thereafter, the optimal thermal and power management methods are introduced.
Nowadays, the chips are generally packed in 2D technique, which means that there is only one layer of dies in the chip. The chip thermal behavior can be described by a 3D heat conduction partial differential equation (PDE). As the target is to balance the thermal behavior and power consumption among the cores, a group of one dimensional (1D) PDEs, which is derived from the developed 3D PDE heat conduction equation, is proposed to describe the thermal behavior of each core. Therefore, the thermal behavior of the MCP is described by a group of 1D PDEs. An optimal controller is designed to manage the power consumption and balance the temperature among the cores based on the proposed 1D model.
3D package is an advanced package technology, which contains at least 2 layers of dies stacked in one chip. Different from 2D package, the cooling system should be installed among the layers to reduce the internal temperature of the chip. In this thesis, the micro-channel liquid cooling system is considered, and the heat transfer character of the micro-channel is analyzed and modeled as an ordinary differential equation (ODE). The dies are discretized to blocks based on the chip layout with each block modeled as a thermal resistance and capacitance (R-C) circuit. Thereafter, the micro-channels are discretized. The thermal behavior of the whole system is modeled as an ODE system. The micro-channel liquid velocity is set according to the workload and the temperature of the dies. Under each velocity, the system can be described as a linear ODE model system and the whole system is a switched linear system. An H-infinity observer is designed to estimate the states. The model predictive control (MPC) method is employed to design the thermal and power management/balancing controller for each submodel.
The models and controllers developed in this thesis are verified by simulation experiments via MATLAB. The IBM cell 8 cores processor and water micro-channel cooling system developed by IBM Research in collaboration with EPFL and ETHZ are employed as the experiment objects.
In system theory, state is a key concept. Here, the word state refers to condition, as in the example Since he went into the hospital, his state of health worsened daily. This colloquial meaning was the starting point for defining the concept of state in system theory. System theory describes the relationship between input X and output Y, that is, between influence and reaction. In system theory, a system is something that shows an observable behavior that may be influenced. Therefore, apart from the system, there must be something else influencing and observing the reaction of the system. This is called the environment of the system.
In recent years, formal property checking has become adopted successfully in industry and is used increasingly to solve the industrial verification tasks. This success results from property checking formulations that are well adapted to specific methodologies. In particular, assertion checking and property checking methodologies based on Bounded Model Checking or related techniques have matured tremendously during the last decade and are well supported by industrial methodologies. This is particularly true for formal property checking of computational System-on-Chip (SoC) modules. This work is based on a SAT-based formulation of property checking called Interval Property Checking (IPC). IPC originates in the Siemens company and is in industrial use since the mid 1990s. IPC handles a special type of safety properties, which specify operations in intervals between abstract starting and ending states. This paves the way for extremely efficient proving procedures. However, there are still two problems in the IPC-based verification methodology flow that reduce the productivity of the methodology and sometimes hamper adoption of IPC. First, IPC may return false counterexamples since its computational bounded circuit model only captures local reachability information, i.e., long-term dependencies may be missed. If this happens, the properties need to be strengthened with reachability invariants in order to rule out the spurious counterexamples. Identifying strong enough invariants is a laborious manual task. Second, a set of properties needs to be formulated manually for each individual design to be verified. This set, however, isn’t re-usable for different designs. This work exploits special features of communication modules in SoCs to solve these problems and to improve the productivity of the IPC methodology flow. First, the work proposes a decomposition-based reachability analysis to solve the problem of identifying reachability information automatically. Second, this work develops a generic, reusable set of properties for protocol compliance verification.
Specification of asynchronous circuit behaviour becomes more complex as the
complexity of today’s System-On-a-Chip (SOC) design increases. This also causes
the Signal Transition Graphs (STGs) – interpreted Petri nets for the specification
of asynchronous circuit behaviour – to become bigger and more complex, which
makes it more difficult, sometimes even impossible, to synthesize an asynchronous
circuit from an STG with a tool like petrify [CKK+96] or CASCADE [BEW00].
It has, therefore, been suggested to decompose the STG as a first step; this
leads to a modular implementation [KWVB03] [KVWB05], which can reduce syn-
thesis effort by possibly avoiding state explosion or by allowing the use of library
elements. A decomposition approach for STGs was presented in [VW02] [KKT93]
[Chu87a]. The decomposition algorithm by Vogler and Wollowski [VW02] is based
on that of Chu [Chu87a] but is much more generally applicable than the one in
[KKT93] [Chu87a], and its correctness has been proved formally in [VW02].
This dissertation begins with Petri net background described in chapter 2.
It starts with a class of Petri nets called a place/transition (P/T) nets. Then
STGs, the subclass of P/T nets, is viewed. Background in net decomposition
is presented in chapter 3. It begins with the structural decomposition of P/T
nets for analysis purposes – liveness and boundedness of the net. Then STG
decomposition for synthesis from [VW02] is described.
The decomposition method from [VW02] still could be improved to deal with
STGs from real applications and to give better decomposition results. Some
improvements for [VW02] to improve decomposition result and increase algorithm
efficiency are discussed in chapter 4. These improvement ideas are suggested in
[KVWB04] and some of them are have been proved formally in [VK04].
The decomposition method from [VW02] is based on net reduction to find
an output block component. A large amount of work has to be done to reduce
an initial specification until the final component is found. This reduction is not
always possible, which causes input initially classified as irrelevant to become
relevant input for the component. But under certain conditions (e.g. if structural
auto-conflicts turn out to be non-dynamic) some of them could be reclassified as
irrelevant. If this is not done, the specifications become unnecessarily large, which
intern leads to unnecessarily large implemented circuits. Instead of reduction, a
new approach, presented in chapter 5, decomposes the original net into structural
components first. An initial output block component is found by composing the
structural components. Then, a final output block component is obtained by net
As we cope with the structure of a net most of the time, it would be useful
to have a structural abstraction of the net. A structural abstraction algorithm
[Kan03] is presented in chapter 6. It can improve the performance in finding an
output block component in most of the cases [War05] [Taw04]. Also, the structure
net is in most cases smaller than the net itself. This increases the efficiency of the
decomposition algorithm because it allows the transitions contained in a node of
the structure graph to be contracted at the same time if the structure graph is
used as internal representation of the net.
Chapter 7 discusses the application of STG decomposition in asynchronous
circuit design. Application to speed independent circuits is discussed first. Af-
ter that 3D circuits synthesized from extended burst mode (XBM) specifications
are discussed. An algorithm for translating STG specifications to XBM specifi-
cations was first suggested by [BEW99]. This algorithm first derives the state
machine from the STG specification, then translates the state machine to XBM
specification. An XBM specification, though it is a state machine, allows some
concurrency. These concurrencies can be translated directly, without deriving
all of the possible states. An algorithm which directly translates STG to XBM
specifications, is presented in chapter 7.3.1. Finally DESI, a tool to decompose
STGs and its decomposition results are presented.
Magnetic spin-based memory technologies are a promising solution to overcome the incoming limits of microelectronics. Nevertheless, the long write latency and high write energy of these memory technologies compared to SRAM make it difficult to use these for fast microprocessor memories, such as L1- Caches. However, the recent advent of the Spin Orbit Torque (SOT) technology changed the story: indeed, it potentially offers a writing speed comparable to SRAM with a much better density as SRAM and an infinite endurance, paving the way to a new paradigm in processor architectures, with introduction of non- volatility in all the levels of the memory hierarchy towards full normally-off and instant-on processors. This paper presents a full design flow, from device to system, allowing to evaluate the potential of SOT for microprocessor cache memories and very encouraging simulation results using this framework.