Computing near the source of data has been proven effective in terms of energy conservation, latency improvement, and privacy preservation. With this, edge intelligence refers to local CNN inference in embedded edge devices. Because edge devices are highly resource-constrained, in practice they store CNN model(s) in external flash memory and load them (or part of) during runtime inference. However, flash memory is subject to time-related retention errors, and thus a dilemma is that without error correction, a CNN model in flash memory quickly deteriorates and becomes useless. On the other hand, strong error correction involves extra read sensing and read retrying. Due to the concern of reliability, the lineup of embedded serial flash memory offers low-density solutions only. In this study, we investigate graceful degradation of CNN models in uncorrected high-density flash storage for edge intelligence. Inspired by the observation that in popular CNN models, the majority of CNN weight parameters share a few monotonic bit patterns in their high-order bits, we propose mapping the frequent bit patterns to low flash-cell voltage levels for protection against retention errors. Furthermore, as CNN layers show different levels of tolerance to bit errors, we also propose using adaptive cell-bit density to provide non-uniform error protection among layers. We conducted experiments on popular CNN designs, including VGG16, ResNet50, and InceptionV3, under realistic effects of flash aging. Results show that with prior methods, the inference accuracy of ResNet50 degrades to 70.7% in just three months; by contrast, with our approach, the degradation is less than 1% and the accuracy remains at 91.6% after one year of retention.

Context-based adaptive arithmetic coding (CAAC) has high coding efficiency and is adopted by the majority of advanced compression algorithms. In this paper, five new techniques are proposed to further improve the performance of CAAC. They make the frequency table (the table used to estimate the probability distribution of data according to the past input) of CAAC converge to the true probability distribution rapidly and hence improve the coding efficiency. Instead of varying only one entry of the frequency table, the proposed range-adjusting scheme adjusts the entries near to the current input value together. With the proposed mutual-learning scheme, the frequency tables of the contexts highly correlated to the current context are also adjusted. The proposed increasingly adjusting step scheme applies a greater adjusting step for recent data. The proposed adaptive initialization scheme uses a proper model to initialize the frequency table. Moreover, a local frequency table is generated according to local information. We perform several simulations on edge-directed prediction-based lossless image compression, coefficient encoding in JPEG, bit plane coding in JPEG 2000, and motion vector residue coding in video compression. All simulations confirm that the proposed techniques can reduce the bit rate and are beneficial for data compression.

The temperature control system has nonlinear time-varying, slow response, time-delay and un-symmetric control input dynamic characteristics. It is difficult to accurately estimate the dynamic model and design a general purpose temperature controller for achieving good control performance. Here a model-free intelligent fuzzy sliding mode control strategy is employed to design a temperature controller with gain-scheduling scheme or auto gain-tuning algorithm for a heating input only closed chamber. The concept of gain scheduling is employed to adjust the mapping ranges of the input and output variables of fuzzy membership functions during the control process for improving the transient and steady-state control performances. The experimental results show that the steady state error of the step input response is always less than 0.2 C without overshoot by using this control schemes.

In this paper, we propose a cross-layer error recovery scheme to counteract channel errors for wireless data communications. It is realized by integrating the multiple reject (MR) ARQ and packet-level coding techniques. Partial error recovery is functioned at the lower layer since the retransmissions are constrained by a timeout value set at the upper layer. The residual errors are expected to be dealt at the upper layer. Packet-level coding exploited at the upper layer may effectively reduce the upper layer retransmissions and thus enhance the system performance. Performance analysis is conducted to demonstrate that our proposed scheme shows substantial benefits especially as the channel becomes noisy.

An effect of multipath on the performance of cellular code division multiple access (CDMA) is investigated. The bit error rate (BER) for both the forward link and reverse link over Rayleigh frequency-selective fading channels are derived. The coherent BPSK modulation with asynchronous direct-sequence (DS) spreading is assumed. The analysis results show the performance degradation of cellular CDMA system as the number of simultaneous users increase. The impact of imperfect power control on the performance of the cellular CDMA system is also investigated.