diff --git a/src/backends/reference/backend.mk b/src/backends/reference/backend.mk
new file mode 100644
index 0000000..23dab11
--- /dev/null
+++ b/src/backends/reference/backend.mk
@@ -0,0 +1,54 @@
+#
+# Copyright © 2017 ARM Ltd. All rights reserved.
+# SPDX-License-Identifier: MIT
+#
+
+# BACKEND_SOURCES contains the list of files to be included
+# in the Android build and it is picked up by the Android.mk
+# file in the root of ArmNN
+
+BACKEND_SOURCES := \
+        RefLayerSupport.cpp \
+        RefWorkloadFactory.cpp \
+        workloads/Activation.cpp \
+        workloads/ArithmeticFunction.cpp \
+        workloads/Broadcast.cpp \
+        workloads/ConvImpl.cpp \
+        workloads/FullyConnected.cpp \
+        workloads/Pooling2d.cpp \
+        workloads/RefActivationFloat32Workload.cpp \
+        workloads/RefActivationUint8Workload.cpp \
+        workloads/RefArithmeticWorkload.cpp \
+        workloads/RefBaseConstantWorkload.cpp \
+        workloads/RefBatchNormalizationFloat32Workload.cpp \
+        workloads/RefBatchNormalizationUint8Workload.cpp \
+        workloads/RefConstantFloat32Workload.cpp \
+        workloads/RefConstantUint8Workload.cpp \
+        workloads/RefConvertFp16ToFp32Workload.cpp \
+        workloads/RefConvertFp32ToFp16Workload.cpp \
+        workloads/RefConvolution2dFloat32Workload.cpp \
+        workloads/RefConvolution2dUint8Workload.cpp \
+        workloads/RefDepthwiseConvolution2dFloat32Workload.cpp \
+        workloads/RefDepthwiseConvolution2dUint8Workload.cpp \
+        workloads/RefFakeQuantizationFloat32Workload.cpp \
+        workloads/RefFloorFloat32Workload.cpp \
+        workloads/RefFullyConnectedFloat32Workload.cpp \
+        workloads/RefFullyConnectedUint8Workload.cpp \
+        workloads/RefL2NormalizationFloat32Workload.cpp \
+        workloads/RefLstmFloat32Workload.cpp \
+        workloads/RefMergerFloat32Workload.cpp \
+        workloads/RefMergerUint8Workload.cpp \
+        workloads/RefNormalizationFloat32Workload.cpp \
+        workloads/RefPermuteWorkload.cpp \
+        workloads/RefPooling2dFloat32Workload.cpp \
+        workloads/RefPooling2dUint8Workload.cpp \
+        workloads/RefReshapeFloat32Workload.cpp \
+        workloads/RefReshapeUint8Workload.cpp \
+        workloads/RefResizeBilinearFloat32Workload.cpp \
+        workloads/RefResizeBilinearUint8Workload.cpp \
+        workloads/RefSoftmaxFloat32Workload.cpp \
+        workloads/RefSoftmaxUint8Workload.cpp \
+        workloads/RefSplitterFloat32Workload.cpp \
+        workloads/RefSplitterUint8Workload.cpp \
+        workloads/ResizeBilinear.cpp \
+        workloads/Softmax.cpp