| Internet-Draft | Abbreviated Title | December 2025 |
| ChazahGroup | Expires 11 June 2026 | [Page] |
The SW103k protocol addresses challenges in networks with limited bandwidth, latency constraints, and data integrity concerns. It provides compression and decompression to optimize bandwidth utilization in environments such as IoT, satellite, and mobile communications.¶
The protocol operates at the data link layer with a custom frame format including SW103K and HAVI headers. Key features include:¶
- Batch processing of 103 packets with compression - Merkle tree-based integrity verification (merklesw103k root hash) - QoS mechanisms with 8-bit priority field - Security features including AES-256-GCM encryption - Physical layer synchronization with +-1us accuracy¶
Implementations include Linux kernel modules, FPGA encoders, and userspace daemons. The protocol supports interoperability with industrial standards like PROFINET and EtherCAT through custom mappings.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 11 June 2026.¶
Copyright (c) 2025 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
This document specifies sw103k, a Layer 2 hybrid compression and decompression protocol designed to achieve extremely high compression ratios for diverse network traffic. sw103k combines template matching, Merkle-based deduplication, Algebraic Sparse Polynomial Compression (ASPC), general-purpose entropy coding (LZ + ANS), and optional machine-learning-based lossy codecs. Deterministic per-frame compression selection ensures predictable behavior with built-in error detection and recovery. The protocol is intended for Layer 2 environments and supports negotiation of compression modes, lossless or lossy operation, and integrity verification. **1. Introduction** This document proposes the SWL103K protocol for interoperable device communication within defined network scopes in order to ensure interoperability **2. Protocol Features** The SWL103K protocol SHOULD support data compression for efficient data exchange in resource-constrained environments. **3. Security Considerations** Implementations of this protocol MUST NOT store plaintext passwords in memory. The rapid growth of networked devices and the emergence of diverse applications have led to the demand for efficient communication protocols that can accommodate varying network conditions, scalability, and resource constraints. The SWL103K protocol presented in this document aims to address these challenges by providing a robust and adaptable solution for data exchange in distributed networks. As network environments become increasingly dynamic and heterogeneous, traditional communication protocols may struggle to provide optimal performance. The SWL103K protocol takes a novel approach by integrating innovative techniques for data transmission, congestion control, and routing. This ensures that the protocol remains responsive and reliable, even in scenarios where network conditions may change unpredictably. This document outlines the fundamental design principles, key features, and operational characteristics of the SWL103K protocol. It describes the protocol's message format, data integrity mechanisms, and how it handles various network scenarios. By providing a comprehensive understanding of the SWL103K protocol, this document aims to enable network engineers, researchers, and implementers to make informed decisions about its adoption and integration into their respective systems. The following sections of this document delve into the specific components of the SWL103K protocol, including its requirements, design considerations, and operational guidelines. Additionally, the document provides insights into its security considerations and interactions with existing protocols. Overall, the SWL103K protocol aims to enhance the reliability, efficiency, and adaptability of communication in modern networked environments What problems does this protocol solve? This protocol solves several problems related to data transmission, compression, decompression, and integrity verification. Specifically, it aims to: Efficiently transmit and manage a large number of small data packets. Compress a batch of 103 data packets into a single compressed data stream. Decompress the compressed data back into the original 103 packets. Calculate and verify the integrity of received data using a merkelsw103k Tree. Handle various states of the communication process, including compression and decompression. +---------------------+ | Incoming L2 Frame | +---------------------+ | v +---------------------+ | Analyzer | | (Entropy/Template) | +---------------------+ | v +---------------------+ | Mode Selector | +---------------------+ | v +---------------------+ | Compressor Engine | | (ASPC, LZ, ML, etc)| +---------------------+ | v +---------------------+ | L2 Header Attachment| +---------------------+ | v +---------------------+ | Compressed Frame | +---------------------+ Modern high-throughput networks increasingly demand compression to reduce bandwidth consumption and improve performance. Layer 2 compression offers the advantage of transparency to higher-layer protocols and reduced latency. Existing approaches (e.g., LZ-based compression, zstd) are limited in compression ratio, adaptability, and cross-format support. sw103k addresses these limitations through a hybrid compression pool capable of adapting per-frame based on content type and entropy. The design goals of sw103k are: High Compression Ratio: Targeting 103:1, adaptable per traffic type. Low Latency: Per-frame compression less than 1 KB to 16 KB frames in under 1 ms. Lossless First, Lossy Optional: Default mode is lossless. Deterministic Selection: Analyzer decisions are consistent at both ends. Cross-Format Support: Supports text, numeric, image, audio, and video streams. Error Detection and Recovery: Mandatory CRC or AEAD integrity checks with fallback support. Protocol requirements are: MUST preserve frame integrity. SHOULD select highest expected compression ratio. MAY employ lossy ML codec if negotiated. MUST provide deterministic per-frame Analyzer decisions. SHOULD provide template and model caching for ratio optimization.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
The provided custom protocol, which appears to be a part of a larger system or application, aims to address various communication and data handling challenges. Below are responses to your questions regarding the abstract understanding of this protocol:¶
| Compression Command C |
|---|
| Decompression Command D Table 1: Compression Modes and Expected Ratios. The sw103k system supports several compression modes with varying expected compression ratios depending on the structure and entropy of the data. Template + Merkle: Used for repeated telemetry or structurally similar frames, with expected ratios between 10:1 and 50:1. ASPC Polynomial: Optimized for numeric matrices, logs, and structured analytics data, producing ratios in the range of 20:1 to 80:1. LZ + ANS: Applied to general text or binary data, typically achieving 2:1 to 10:1 compression. Transform Text: Intended for structured or repetitive text, with expected ratios of 5:1 to 15:1. ML Autoencoder: A lossy mode for audio, video, or images, targeting compression ratios up to 103:1. Passthrough: Used for high-entropy data where compression is not effective, producing a 1:1 ratio. Protocol Overview – Text Only sw103k operates as a Layer 2 shim between the MAC and network interface. It analyzes each frame and selects the most efficient compression algorithm. The Analyzer performs entropy estimation, template matching, polynomial fitting, and ML model inference. Frames are compressed using the selected mode and wrapped with a minimum 12-byte header. Frame Pipeline Overview: Frame → Analyzer → Mode Selector → Compressor Engine → Header Generator → Compressed Frame. Frame Format – Text Only The sw103k header is a minimum of 12 bytes and is TLV-extensible. The header includes: version, algorithm (ALG), lossy flag, estimated ratio, model identifier, flags, original length, compressed length, and integrity tag (CRC or AEAD). Header Layout (text description): Version and algorithm fields occupy the first bits. Lossy flag and estimated ratio follow. A model identifier and flags are next. Two length fields represent original and compressed sizes. A 32-bit or larger checksum or AEAD authentication tag completes the header. ALG Code Points: 0x0 = Passthrough 0x1 = Template + Merkle 0x2 = ASPC Polynomial 0x3 = LZ + ANS 0x4 = Transform Text 0x5 = ML Autoencoder (Lossy) Compression Modes – Text Only Template + Merkle Deduplication Uses pre-shared templates or a sliding window of recent frames. Each frame is divided into chunks; if a chunk hash matches the Merkle tree, a short identifier is emitted instead of raw data. ASPC Polynomial Compression Numeric or log data is approximated using sparse polynomials. Only coefficients and residuals are stored. Example: values [100,102,104,106] fit a linear polynomial, producing negligible residuals and very small compressed size. LZ + ANS Traditional dictionary compression (LZ77) followed by asymmetric numeral system entropy encoding. Transform Text Uses BWT, Move-to-Front, and ANS to compress structured textual content. ML Autoencoder (Lossy) Domain-trained autoencoders compress audio, video, or images with lossy reconstruction. Ratios may reach 103:1 depending on complexity and quality settings. Passthrough High entropy or encrypted content is forwarded without compression. Analyzer FSM – Text Only The Analyzer determines compression mode based on frame entropy: If entropy is below a low threshold, template or polynomial modes are selected. If entropy is moderate, LZ or Transform modes are used. If entropy is high but lossy compression is allowed, the ML autoencoder is selected. If entropy is too high or lossy is disallowed, the frame is passed through unchanged. Flow Summary: Frame → Entropy Estimate → (Low → Template/ASPC; Medium → LZ/Transform; High → ML if allowed; otherwise Passthrough) Negotiation – Text Only During link setup, both peers exchange TLVs describing supported compression modes, latency limits, maximum dictionary size, AEAD requirements, and model identifiers. After capability exchange, the link transitions to operational mode. Error Detection and Recovery – Text Only Integrity is provided using CRC32/64 or AEAD tags. On decompression failure, peers request retransmission or updated templates/models. Persistent errors cause fallback to passthrough mode. Security Considerations – Text Only Compression-before-encryption is recommended for optimal ratios. Lossy mode is allowed only if negotiated. Integrity checks prevent silent corruption. Entropy analysis must not leak sensitive information. Operational Considerations – Text Only Hardware acceleration is recommended for real-time operation. Template and model caching significantly improves compression efficiency. Operators may adjust Analyzer thresholds and enable telemetry for optimizing compression behavior. IANA Considerations – Text Only New registries are required for sw103k algorithm (ALG) code points, TLV types, and model identifier versions. Appendix A – Text-Only Example Example compression: Original frame: 1024 bytes telemetry Mode: ASPC polynomial Compressed payload: 120 bytes Header: 9 bytes CRC: 4 bytes Total: 133 bytes Compression Ratio ≈ 7.7:1 Appendix B – Merkle Template Example A set of eight recent frame hashes is stored along with 32-bit short-ID mappings. A Merkle root references all recent chunks, enabling lightweight template reuse across frames. Appendix C – Pseudocode Summary If entropy is low, use template or ASPC. If entropy is medium, use LZ or Transform. If entropy is high and lossy is allowed, use ML Autoencoder. Otherwise, use passthrough. |
<CODE BEGINS> file "network_app_protocol.c"
<CODE BEGINS>
#include "network_app_protocol.h"
int main() {
// Initialize your custom protocol and
// perform any necessary setup
struct sw103k_proto mp;
// Initialize mp and set its initial
// state, buffers, etc.
// Example function calls
sendCommand(&mp, "CONNECT");
authenticate(&mp, "networkuser",
"networkpassword");
// Continuously receive and process data
while (1) {
custom_receive(&mp, network_socket);
// Replace 'network_socket' with your
// actual socket
}
// Clean up and exit
// Close sockets, free memory, etc.
return 0;
}
<CODE ENDS>
<CODE ENDS>
This memo includes no request to IANA.¶
This document should not affect the security of the Internet.¶
Appendix¶
This work is supported by Chazah Group¶
Thanks to Chazah Group Ltd¶
Changes in this version (draft-rfcxml-rfc-swl-103k-03):¶
- Clarified the applicability of the SWL103K protocol to avoid implying¶
global deployment.¶
- Updated normative language: changed "MUST be implemented by all network¶
devices" to a more realistic recommendation for scoped environments.¶
- Responded to feedback from Area Directors regarding clarity and protocol¶
deployment assumptions.¶