As heterogeneous systems scale, traditional rule-based and stochastic methods used for chip placement face limitations in convergence, scalability, and thermal management. To address these challenges, we propose a reinforcement learning (RL) framework with multi-objective optimization, employing a customized reward shaping method to minimize interconnect wirelength and improve thermal distribution. We applied our approach to two generic use cases in 2.5D advanced packaging - a 4-chiplet RDL system and a multi-GPU system. Our results outperformed the state-of-the-art methods like Bayesian optimization (BO) by up to \(40\%\) in wirelength reduction with 4°C thermal improvements.

The placement of chiplets on a silicon interposer is a pivotal step in 2.5D system integration, yet existing placement approaches typically assume a pre-defined interposer footprint. This creates a circular dependency: the optimal footprint cannot be known without first solving the placement, while the placement itself is constrained by the given dimensions. An undersized interposer may exclude feasible placements, while an oversized one yields unnecessarily sparse solutions. Moreover, even when the footprint area is minimized, few existing approaches explicitly control the interposer’s aspect ratio. To jointly address these challenges, we propose FAPlace, a footprint aware mask guided sequential placement framework. FAPlace operates on a sufficiently large canvas, eliminating the circular dependency by allowing the optimal interposer footprint to emerge as an output of the optimization rather than a pre-specified input. At its core is a novel footprint mask that fuses area compactness with an aspect ratio penalty into a unified spatial cost map. Integrated with wirelength and thermal guidance masks, FAPlace delivers holistic multi-physics optimization in a deterministic, single pass process. Experimental results demonstrate that FAPlace reduces wirelength and footprint area while achieving near-unity aspect ratios, without compromising on thermal performance.

Active learning, which aims to enhance modeling efficiency, precision, and cost effectiveness through selective labeling, is emerging as a promising strategy for analog circuit modeling. However, analog circuits are constrained by strict functional and technological limitations, resulting in scarcity of data for modeling, and additional data acquisition involves expensive and time-consuming simulations. For efficient and effective active learning for analog circuit modeling, our research analyzes data-driven initial sampling techniques which lays the foundation for the active learning process. Our experiments reveal that these initialization strategies expedite the learning process, decrease the demand for extensive simulations, and produces more accurate models. Furthermore, the results demonstrate that active learning techniques, which uniformly sample the design space, tend to benefit from distance-based initialization technique.

Computationally intensive simulations have made analog circuit sizing challenging for complicated analog circuit performance characterization. Accurate yet computationally efficient data-driven models of circuit performance can potentially accelerate the design and verification process. However, as analog circuits are designed under strict functional and technology constraints, there is a scarcity of data for analog circuit performance classification, posing challenges to data-driven approaches; acquiring more data typically involves running expensive and time consuming simulations. We propose Bayesian Deep Active Learning (BDAL) to learn models using fewer simulations, by iteratively selecting a small number of informative samples to label based on the model uncertainty. Bayesian neural networks used in the BDAL framework are better able to model weight uncertainty while being sufficiently expressive to model complex circuits. Compared with the state-of-the-art approaches, the proposed BDAL method can obtain better classification performance with much fewer number of simulations. Experiments on four diverse analog circuits demonstrate BDAL can achieve significant reduction in data requirement and obtain similar performance with much less labeled data for analog circuit performance classification.

We propose a three-track detection system for two dimensional magnetic recording (TDMR) in which a local area influence probabilistic (LAIP) detector works with a trellis-based Bahl-Cocke-Jelinek-Raviv (BCJR) detector to remove intersymbol interference (ISI) and intertrack interference (ITI) among coded data bits as well as media noise due to magnetic grain-bit interactions. Two minimum mean-squared error (MMSE) linear equalizers with different response targets are employed before the LAIP and BCJR detectors. The LAIP detector considers local grain-bit interactions and passes coded bit log-likelihood ratios (LLRs) to the channel decoder, whose output LLRs serve as a priori information to the BCJR detector, which is followed by a second channel decoding pass. Simulation results under 1-shot decoding on a grain-flipping-probability (GFP) media model show that the proposed LAIP/BCJR detection system achieves density gains of 10.16% for center-track detection and 3.13% for three-track detection compared to a standard BCJR/1D-PDNP. The proposed system's BCJR detector bit error rates (BERs) are lower than those of a recently proposed two-track BCJR/2D-PDNP system by factors of (0.55, 0.08) for tracks 1 and 2 respectively.

The rapid growth of commercial cellular and personal communication systems has increased the demand for high data rate, power and bandwidth efficient transmissions. Advanced coding and modulation techniques are suitable candidates for this purpose as the potential to reduce the symbol period is often limited by the multipath-induced intersymbol interference (ISI). In this paper, a scheme that jointly performs detection, demodulation, and decoding is proposed and is coined the joint detector demodulator decoder (JDDD). The JDDD is a generalization of the previously proposed joint Viterbi detector decoder (JVDD) that performs detection and decoding of binary phase shift keying (BPSK) modulated signals over ISI channels. The JVDD algorithm performs two operations on a trellis: the first operation computes metrics and retains a set of survivors, while the second operation performs parity checking on the survivors to return the minimum metric legal codeword (MMLC). The JDDD algorithm also has a similar structure but the number of states in JDDD trellis is a factor of the number of symbols defined by the modulator and the characteristics of the ISI channel. In this paper, the performance of the proposed JDDD algorithm is evaluated over an ISI channel.

Multihop cooperative communication is emerging as a key concept to extend the coverage area of the network and potentially increase the capacity. The spectral efficiency of such networks can be improved by adapting the transmission to time-varying channel conditions, referred to as incremental relaying. Although such incremental relaying concepts are progressively being studied, many challenges, such as erroneous transmissions by intermediate nodes and end-to-end delay of the network, limit its practical use due to lack of an efficient implementation. This paper proposes an efficient multihop incremental relaying technique. In this method, erroneous relay forwarding is mitigated, and the overhead for coordination among nodes is reduced by exploiting the implicit feedback channel available due to the broadcast nature of wireless transmissions. The proposed scheme fully leverages the benefit of overhearing and eliminates the additional feedback slots required for validation. Further, it ensures reliable forwarding of information, which optimizes the throughput of multihop networks. Thorough analysis of the proposed scheme is performed under different deployment environments, and the theoretical analyses presented in this paper are supported with results from extensive simulation studies.

Rateless code properties ideally suit multi-node transmissions and have been extensively employed in multihop delay tolerant networks (DTNs). However conventional multihop networks, where the transmission proceeds from one-hop to the next, may lead to starvation of some nodes with good channel conditions. In such instances, the overall latency can be alleviated by cooperation among the forwarding nodes. This paper analytically quantifies the latency of both conventional and cooperative rateless coded delay tolerant multihop networks by deriving the exact closed form equations for the channel usage. These are verified and supported with results from extensive simulation studies.

Multihop cooperative communication is emerging as a key concept to extend the coverage area and potentially increase the capacity of wireless networks. With multiple hops, processing at the forwarding nodes is a major source of error propagation. In this context, a novel spectrally efficient transmission scheme with error protection for such networks is proposed in this paper. The proposed scheme retains the spatial diversity and enhances the bandwidth efficiency by integrating incremental redundancy concepts into such cooperative multihop systems. The guard against error propagation and the overhead for coordination among multiple nodes in such networks is mitigated by the implicit feedback channel available due to broadcast nature of wireless transmissions. The proposed scheme fully leverages the benefit of overhearing and eliminates the additional feedback slots required for validation. Further, it ensures reliable forwarding of information which optimizes the throughput of multihop networks. Theoretical framework is also developed for the proposed system and supported with results from extensive simulation studies.

Rateless codes have been found to be particularly attractive for decode and forward based relaying strategy in delay tolerant multihop networks. The latency performance of such networks is dependent on the worst links that results in the starvation of subsequent nodes with good channel conditions. The total delay suffered by such networks can be constrained by limiting the number of rateless coded transmissions, especially for applications with critical latency requirements. However, the performance of rateless codes deteriorates in such circumstances due to the lack of sufficient mutual information for successfully recovering the entire source packets. The fraction of source packets that can be recovered will depend on the encoded packets received across the transmission channel. This paper investigates the degradation in the performance of such rateless coded networks by deriving the average packet recovery rate. In order to improve the reliability in such delay constrained networks, a novel spectrally efficient transmission scheme for reliable multihop data transfer is proposed. The proposed scheme exploits the broadcast nature of wireless transmissions, which provides an inherent implicit feedback channel, to ensure the reliable delivery of information packets to the nodes in the network. Rather than allocating dedicated channels to feedback the packet recovery information, the implicit feedback channel determines such information which enhances the spectral efficiency. Further, the optimum number of packets recoverable within the specified delay constraint to reduce the reprocessing of lost packets across hops is analytically analysed in this paper.

Cooperation among nodes is proposed as an effective means of combating fading and for enhancing system's overall capacity and coverage. The robustness of rateless codes makes them particularly attractive for such networks. Yet, the decoding aspects of rateless codes are discarded in most of the traditional distributed networks. In this paper, the intermediate packet decodability of rateless codes is exploited to reduce the number of packets being processed at the relay nodes. This is achieved by harnessing the back channel (from destination to relays) for feedback. The relay transmissions and processing are thereby confined to assist the receiver in decoding the remaining information. It is shown both analytically and through simulation studies that such a scheme achieves significant savings in computation complexity, memory usage and overall energy consumption.

When a message is forwarded through a network, a node will have multiple opportunities to reliably receive the message and cooperate in transmission. However, the transmit power of each node in a network is limited and so is the transmission range. Therefore, multihop transmission is a potential technique to deliver the data over large distances. Rateless code properties ideally suit decode and forward based relaying strategies as it improves the outage and bandwidth efficiency. In decode and forward based multihop networks, the messages are decoded and relayed over sequential point-to-point communication links. In a conventional multihop network, the intermediate nodes will have to wait indefinitely to correctly decode the information. To quantify the impact of this unknown waiting time, this paper analyses the performance of a multihop network with strict delay constraints. The average packet loss for such a network is obtained as a function of the total delay. By optimizing the total delay based on the performance across each links, the average packet loss can be minimized.

Cooperation among nodes in a network provides spatial diversity as well as it enhances the system's overall capacity and coverage. However the benefits provided by such systems depends on the level of cooperation (based on the relay information quality) and on the coordination among nodes. One way of achieving proper coordination is through the use of feedback channels. An implicit feedback channel available during relaying has been determined to minimize the feedback among cooperating nodes. In addition, it improves the reliability of information decoding at the destination by mitigating erroneous relay transmissions. The present paper studies the outage analysis of a cooperative transmission scheme assisted by the implicit feedback channel. The analysis is performed in a delay constrained network employing rateless codes. Further, a performance improvement of the order of 2.5 dB is observed for such networks.

In wireless networks, designing transmission schemes adapting to time varying channel conditions is key to achieve better spectral efficiency. The design of such schemes for multi-hop wireless networks requires coordination among nodes in the network, which is achieved through feedback between nodes. Spectral efficiency can be improved for such networks by designing transmission schemes with limited feedback. Incremental code relaying (ICR) with implicit feedback proposed previously is one that mitigates the use of feedback link. In this paper, location-aware ICR with power allocation is proposed that dynamically makes the routing decisions based on the link that offers the maximum capacity. Theoretical upper bounds for bit-error rate (BER) is derived and is verified by the simulation results, yielding further insight to the proposed scheme.

In this paper, an incremental redundancy based cooperative coding scheme for wireless networks is proposed. User cooperation is an effective way of mitigating fading where one user cooperates with another user in transmitting its information, thereby providing spatial diversity. Wireless transmissions are broadcast by nature and this provides a new dimension to the system in terms of an implicit feedback channel from the relay to the source. To exploit the implicit feedback channel a novel incremental redundancy based transmission scheme called incremental code relaying with implicit feedback (ICRIF) is proposed. This scheme enhances the reliability of information transmitted by the relay and thereby improving the throughput at the destination. By employing a powerful code such as low-density parity-check (LDPC) code, the benefit of the proposed scheme is clearly established. The simplified error probability for correct reception of the proposed scheme is derived with respect to direct transmissions and transmissions assisted by the relay.

In this paper, the application of rate-compatible punctured convolutional (RCPC) code into coded cooperation relay system is studied. As opposed to conventional coded cooperation systems, RCPC is attractive for its better spectral efficiency. In addition, the broadcast nature of relay transmission provides a new dimension to the system in the form of implicit feedback channel from relay to source. This allows the source node to justify the quality of information bits at relay node and react accordingly by taking over relay's role to transmit the additional parity. At the destination, selection code combining is then performed, which provides a diversity benefit to the underlying RCPC code. The performance of the proposed scheme in terms of the number of correctly decoded bits (goodput), bit error probability, and the required number of slots for message transmission are then analysed. Extension to the case where there are multiple relays to assist the transmission is also discussed.