High-Bandwidth Low-Latency
Networks for Aerospace
Under the working title “Aero/Spine” MLE provides network infrastructure based on open standards such as IEEE Ethernet Time-Sensitive Networking (TSN) and focused on Aerospace/Defense system.
Missing Link Electronics (MLE) provides lossless, gapless network and storage connectivity and cross-plane interoperability across TCP/UDP/IPv4, PCI Express (PCIe), NVM Express (NVMe), and high-speed Ethernet for modern MOSA/SOSA-aligned aerospace systems. By offloading complex protocol processing entirely into FPGA programmable logic, MLE eliminates the host CPU bottlenecks, operating system jitter, and packet drop risks inherent in traditional software stacks—providing the deterministic data throughput required for mission-critical electronic warfare (EW), radar, SIGINT, and high-resolution ISR payloads. MLE’s Aero/Spine Solutions run on trusted hardware solutions from partners like Alpha-Data.
Features
- High-speed analog and digital data acquisition
- Lossless and gapless recording of sensor data
- Interface to 1 / 2.5 / 5 / 10 / 25 / 40 / 50 / 100 / 200 / 400 Gigabit Ethernet
- Full-duplex with 128 bit wide bidirectional datapath
- Full line rate of 70 Gbps, or more, per instance in FPGA
- Low one-way latency NPAP-to-NPAP (600 nanoseconds for 100 Bytes)
- Network diagnostics functions (optional)
- TCP session priority management (optional)
- Transport Layer Security (TLS) (optional)
- Time-Sensitive Networking (TSN) (optional)
- Network Impairment Generators (optional)
Applications
- RF Signal Sampling/Generation
- SOSA Systems: 14.6.11 Slot Profile
- Electromagnetic Spectrum Operations (EMSO)
- Radar/Sensor Recording
- Military Communication (MilCom)
- RF Signal Sampling/Generation
- Beamforming
- Signal Detection/Jamming
- Pre-6G Systems
Core Solution Capabilities
High-Throughput Data Plane Acceleration (10G / 25G / 100G+)
Supports line-rate, hardware-accelerated UDP/IPv4 and deterministic TCP/IPv4 transport over 25GBASE-KR and 100GBASE-KR4 OpenVPX backplanes. Purpose-built for streaming raw digitized sensor data, high-resolution video, and VITA 49 (VRT) radio transport packets with zero dropped frames.
Gapless Sensor-to-Storage Recording (PCIe & NVMe Expansion Plane)
Combines high-bandwidth PCIe (Gen 3/4/5 up to x16) with dedicated NVMe storage controller IP. Enables direct, continuous streaming from high-speed ADCs/DACs directly to NVMe SSD arrays at multi-gigabyte line rates, completely bypassing host CPU memory buses during sustained sensor bursts.
Deterministic, Zero-Jitter RTL Execution
Implements full protocol processing as standalone hardware state machines inside the FPGA fabric. By bypassing standard OS kernel drivers, the architecture guarantees fixed, microsecond-level latency and predictable synchronization across distributed backplane slots.
SWaP-C Optimization & System Hardening
Offloading heavy transport, packet parsing, and NVMe driver tasks into compact FPGA logic drastically reduces thermal dissipation, power draw, and physical footprint. Simultaneously, the hardware-isolated, air-gapped pipeline removes OS attack surfaces to resist protocol-level cyber threats.
Turnkey Interoperability Across COTS & Custom Hardware
Fully compatible with standard 3U/6U OpenVPX (VITA 65) slot profiles on commercial off-the-shelf (COTS) SOSA-aligned FPGA platforms (including AMD UltraScale+ and Altera Agilex) as well as proprietary designs, backed by comprehensive end-to-end design and system integration services.
Radar and Sonar Signal Processing & Recording
MLE Aero/Spine Solutions accelerates radar and sonar response analysis and streaming by providing seamless data input processing. From high-performance data capture and processing, beamforming and switching, to data streaming and recording, MLE’s solutions enables faster analysis and high-speed transfer with reliability.
Video and Image Acquisition and Processing
MLE micro-second connectivity delivers high-performance, ultra-low-latency image processing and data streaming with industry standard camera interfaces for lossless and gapless frame or video stream capture, processing and recording.
Rugged and Embedded Communications
MLE Aero/Spine Solutions accelerate aerospace system communication in harsh environments. From high-performance data processing to data switching, the solutions comprehensively manage the data path, configuring the PCIe core for high-speed transfer for communications and radar applications.
High-Bandwidth Low-Latency Connectivity
Conventional software stacks and standard storage interfaces create latency jitter, host CPU overhead, and security risks that aerospace platforms cannot afford. MLE supports this with fully hardware-accelerated TCP/UDP/IPv4, PCIe, NVMe, Ethernet, or proprietary protocols running entirely in FPGA logic—completely independent of the OS. Deployed on MOSA/SOSA-aligned hardware, MLE ensures deterministic, zero-jitter line-rate throughput from 10G to 100G+, significantly accelerating deployment timelines while slashing integration costs and SWaP-C.
SOSA-Aligned Development Kit for Aerospace Systems
MLE and partners supports many off-the-shelf MOSA/SOSA-aligned FPGA boards and custom FPGA boards with design integration services:
MLE NVMePL-x4 on Alpha Data ADA-VA330 Board
 FFSS-ADA-VA330-NVMePL-x4 Block Diagram | MLE NVMePL-x4 on Alpha Data ADA-VA330 BoardMLE NVMePL-x4 is a full accelerator for NVMe data streaming with 4 lanes in PL based on MLE NVMe Fast FPGA RAID (NVMe FFRAID). Alpha Data’s ADA-VA330 board is an 14.6.11 SOSA-Aligned 3U VPX Versal™ RF module with AMD Versal™ RF VR1602. The FFSS-ADA-VA330-NVMePL-x4 combines MLE’s NVMe FFRAID Subsystems with Alpha Data’s ADA-VA330 board featuring the AMD Versal™ RF VR1602 for 16 DACs and 16 ADCs at 14-bit resolution with up to 18GHz bandwidth combined with powerful hardened IP for FFT, channelization, LDPC decode, and resample. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-VA330-NVMePL-x4 comprehensively manages the data path, configuring the PCIe core for high-speed transfer. |
MLE NVMePL-x1 on Alpha Data ADA-VA330 Board
FFSS-ADA-VA330-NVMePL-x1 Block Diagram | MLE NVMePL-x1 on Alpha Data ADA-VA330 BoardMLE NVMePL-x1 is a full accelerator for NVMe data streaming with 1 lane in PL based on MLE NVMe Fast FPGA RAID (NVMe FFRAID). Alpha Data’s ADA-VA330 board is an 14.6.11 SOSA-Aligned 3U VPX Versal™ RF module with AMD Versal™ RF VR1602. The FFSS-ADA-VA330-NVMePL-x1 combines MLE’s NVMe FFRAID Subsystems with Alpha Data’s ADA-VA330 board featuring the AMD Versal™ RF VR1602 for 16 DACs and 16 ADCs at 14-bit resolution with up to 18GHz bandwidth combined with powerful hardened IP for FFT, channelization, LDPC decode, and resample. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-VA330-NVMePL-x4 comprehensively manages the data path, configuring the PCIe core for high-speed transfer. |
MLE NVMePS on Alpha Data ADA-VA330 Board
 FFSS-ADA-VA330-NVMePS Block Diagram | MLE NVMePS on Alpha Data ADA-VA330 BoardMLE NVMePS is a storage stack to read/write data onto a Linux-connected NVMe SSD using Linux file system. Alpha Data’s ADA-VA330 board is an 14.6.11 SOSA-Aligned 3U VPX Versal™ RF module with AMD Versal™ RF VR1602. The FFSS-ADA-VA330-NVMePS combines MLE’s Linux-based storage stack with Alpha Data’s ADA-VA330 board featuring the AMD Versal™ RF VR1602 for 16 DACs and 16 ADCs at 14-bit resolution with up to 18GHz bandwidth combined with powerful hardened IP for FFT, channelization, LDPC decode, and resample. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-VA330-NVMePS manages the data path, configuring the PCIe core for high-speed transfer. |
MLE NVMePL-x4 on Alpha Data ADA-V9202 Board
 FFSS-ADA-V9202-NVMePL-x4 Block Diagram | MLE NVMePL-x4 on Alpha Data ADA-V9202 BoardMLE NVMePL-x4 is a full accelerator for NVMe data streaming with 4 lanes in PL based on MLE NVMe Fast FPGA RAID (NVMe FFRAID). Alpha Data’s ADA-V9202 board is a SOSA-Aligned 3U VPX module with AMD Zynq® Ultrascale+ XCZU47DR-2 or XCZU48DR-2 (FFVE1156). The FFSS-ADA-V9202-NVMePL-x4 combines MLE’s NVMe FFRAID Subsystems with Alpha Data’s ADA-V9202 rugged SOSA-Aligned 3U VPX format module and is available in industrial temperature grades with conduction cooling. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-V9202-NVMePL-x4 comprehensively manages the data path, configuring the PCIe core for high-speed transfer. |
MLE NVMePL-x1 on Alpha Data ADA-V9202 Board
FFSS-ADA-V9202-NVMePL-x1 Block Diagram | MLE NVMePL-x1 on Alpha Data ADA-V9202 BoardMLE NVMePL-x1 is a full accelerator for NVMe data streaming with 1 lane in PL based on MLE NVMe Fast FPGA RAID (NVMe FFRAID). Alpha Data’s ADA-V9202 board is a SOSA-Aligned 3U VPX module with AMD Zynq® Ultrascale+ XCZU47DR-2 or XCZU48DR-2 (FFVE1156). The FFSS-ADA-V9202-NVMePL-x1 combines MLE’s NVMe FFRAID Subsystems with Alpha Data’s ADA-V9202 rugged SOSA-Aligned 3U VPX format module and is available in industrial temperature grades with conduction cooling. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-V9202-NVMePL-x1 comprehensively manages the data path, configuring the PCIe core for high-speed transfer. |
MLE NVMePS on Alpha Data ADA-V9202 Board
 FFSS-ADA-V9202-NVMePS Block Diagram | MLE NVMePS on Alpha Data ADA-V9202 BoardMLE NVMePS is a storage stack to read/write data onto a Linux-connected NVMe SSD using Linux file system. Alpha Data’s ADA-V9202 board is a SOSA-Aligned 3U VPX module with AMD Zynq® Ultrascale+ XCZU47DR-2 or XCZU48DR-2 (FFVE1156). The FFSS-ADA-V9202-NVMS combines MLE’s Linux-based storage stack with Alpha Data’s ADA-V9202 rugged SOSA-Aligned 3U VPX format module and is available in industrial temperature grades with conduction cooling. All key functionality is pre-integrated and pre-validated and ships as a compilable FPGA design project along with an IP-Core license from MLE. The FFSS-ADA-V9202-NVMePS manages the data path, configuring the PCIe core for high-speed transfer. |