Dynamic-Coding-Based SMC with Multi-Node Protocol under Gilbert-Elliott Channel: An Intelligent Bit-Rate Pre-Allocation Method

This paper is concerned with a dynamic-coding-based sliding mode control and intelligent bit-rate pre-allocation method for networked control systems under Gilbert-Elliott channels. To alleviate limited communication bandwidth and packet-length-dependent packet dropout, a channel-guided multi-node round-robin-like protocol is proposed. This protocol allows the sensor side to utilize previous channel mode information to dynamically determine the number of accessing nodes, thereby effectively preventing a surge in packet dropout probability when network conditions deteriorate. Furthermore, a dynamic quantization encoding-decoding mechanism is designed to encapsulate channel modes, zoomed state to be quantized, and dynamic quantization parameters into a binary packet, while an input-holding strategy at the receiver precisely compensates for potential state losses. Crucially, to maximize steady-state control precision, the minimization of the convergent upper bound is transformed into a mixed-integer programming problem. An intelligent optimization scheme combining the simulated annealing algorithm with linear matrix inequalities is proposed to optimally allocate bit rates among different nodes subject to total bandwidth constraints. Theoretical analysis proves that the closed-loop system achieves exponential ultimate boundedness in mean square and ensures the reachability of the sliding domain. Finally, simulation results verify the effectiveness of the proposed strategy.